Abstract: The subject of research is the steganographic method of embedding information in digital images. Steganography is capable of hiding not only the content of information, but also the very fact of its existence. The paper considers one of the most important problems in the development of steganographic methods - the secrecy of the transfer of protected information. Stealth is not only visual or auditory indistinguishability of a digital media resource from a media resource with embedded information, but also statistical indistinguishability. Special attention is paid to preserving the spatial statistical dependence between the image pixels. The methodological basis of the research is the methods of mathematical statistics and image processing theory, as well as image distortion metrics. The novelty of the research lies in the development of a new method of embedding information in static images. The authors consider in detail the problem of applying the moving average autoregression process to represent the statistical dependence of image pixels. It is shown that the proposed method allows you to embed information into digital images without significant distortion.

Abstract: The research subject is the steganographic methods of data protection, able to hide not only the contents of the transmitted information, but also the very fact of transmission. The authors consider an important problem of creation of steganographic systems of information protection – the high secrecy of information transmission, which is the most important characteristic of such systems. The secrecy is understood not only as a visual or sound indistinguishability of a digital media file from the media file with embedded information, but also the statistical indistinguishability. The research methodology is based on the objective metric of image distortion, including the generally accepted figure – peak signal to noise ratio. The scientific novelty of the study consists in the proposed method of increase of the hidden transmission of information using the minimal possible image distortion with its maximal filling with embedded data. The authors analyze the most widely used steganographic systems of information protection. The authors ascertain that the most promising method of steganography is the method of direct spectrum broadening, since it has particular advantages compared with other methods. The problem of a high level of secrecy of information transmission, connected with the unacceptably high level of a container distortion upon the hidden data embedding is solved using this method. Based on the peak signal to noise ratio, the authors study the distortion of digital images with hidden information, embedded using different methods. The proposed method demonstrates the best results. 

Abstract: The subject of the research is the steganographic information protection system (SIPS) based on the method of the direct spectrum spread with minimum change of image pixels and maximum image fill. Special attention is paid to the problem of creating a SIPS with a blind decoder because the original image strongly influences the correct extraction of in-built information. The authors examine such aspects of the topic as the application of the autoregression process and error-correcting codes in order to reduce the number of errors during extraction of in-built information. The methodological basis of the research involves methods from the theory of noise-resistant coding, methods of digital image processing, and methods of mathematical statistics. The novelty of the research is caused by the fact that the author defines the optimum block size (32x32 pixels) that enables the two-dimensional autoregression process (AP) to provide the maximum number of correctly extracted in-built bits. Moreover, the authors analyze efficiency of applying BCH Code, Reed-Solomon code and LDPC code for restoration of in-built data. They demonstrate that LDPC code combined with the algorithm of 'soft' decision making 'Belief propagation' option is the most efficient for reducing the number of errors when extracting in-built information by the stenographic method on the basis of the direct spectrum spread.